1. Levels and Predictors of LHDs' Engagement in Community Health Assessment, 2002-2013 Gulzar Shah, PhD, MS Associate Dean for Research Associate Professor, JPHCOPH Georgia Southern University Kay Lovelace, PhD, MPH Associate Professor, Department of Public Health Education The University of North Carolina at Greensboro Daniel Linder, PhD Assistant Professor, JPHCOPH Georgia Southern University

2. Disclosure Neither we, nor our spouses or partners have had a financial, professional or personal relationship that might potentially bias and/or impact the content of the educational activity/session. Gulzar Shah, Kay Lovelace, Daniel Linder

3. Community health assessments are.. A cross-cutting element of the public health infrastructure One of three prerequisites for entering the PH accreditation process Required of non-profit hospitals, with public health input, under the PPACA Integral to the community health improvement process

4. Little is known about how CHAs are used to... Adopt and implement evidence-based interventions, policies, environmental and systems change. To study this, we first wanted to know the extent to which and what enables LHDs to engage in Community Health Assessment processes.

5. Our questions: What is the extent of LHDs’ use of CHA/CHIP processes from 2002- 2013? What factors predict that use?

6. Methods: Dataset/Sample NACCHO Profile of LHDs Study –2005 –2008 –2010 –2013 All LHDs completing the Profile for these years – longitudinal study Descriptive/Bivariate: –1378 LHDs (55% of all LHDs) Multivariate Analysis: –Data for all four points in time combined (appended); –Used year of Profile Study as a variable.

7. Methods: Variables Dependent Variable for multivariate analysis Completion of CHA within the last 5 years by LHDs (drawn from 2005, 2008, 2010, 2013) Independent Variables Jurisdiction population Personnel Governance Year of data collection

8. What is the extent of LHDs’ use of CHA from 2002-2013? LHDs completing CHAs during 4 NACCHO Profile Studies

9. Methods: Multivariate Analysis Generalized linear mixed model –Used to model associations between variables when the outcome of interest is correlated among observations. –We fit a generalized linear mixed model with a random intercept to account for the correlated nature of the dependent variable (CHA) over time.

10. Results: AORs

11. Which factors predict the use of CHA over time? Bottom line: Jurisdiction size Personnel State / local governance relationship

12. LHDS in larger jurisdictions are more likely to complete CHAs

13. LHDs with Health Educators are more likely to complete CHAs EpidemiologistHealth Educator

14. State & shared governance LHDs were less likely to complete CHAs

15. AOR: Year of Data Collection

16. Discussion

17. Population size –Economies of scale and scope may lead to higher rates of CHA completion Centrally governed LHDs –LHDs may have less autonomy for decision making –States may conduct CHAs for LHDs

18. Personnel –Health educators lead CHA process in many states. –The finding regarding epidemiologists is unexpected. AOR is relatively flat from year to year. –Perhaps accreditation is not driving completion of the CHAs.

19. CHAs are resource intensive. What do LHDs gain from completing a quality Community Health Assessment? How do LHDs obtain/share resources for CHA? With what partners do LHDs collaborate? How do LHDs and their partners use the data from CHAs? What difference does completing a CHA make in the health of a community?

20. Questions? Gulzar Shah gshah@georgiasouthern.edu Kay Lovelace kalovela@uncg.edu